So Long, Jobless Recovery

Increasing demand and smaller productivity gains mean payrolls will grow -- perhaps at a 200,000 jobs-per-month rate -- into 2004

After two long years, the recovery from the 2001 recession is no longer jobless. The Labor Dept. says payrolls have risen for three consecutive months, although at a pace well below that of a typical recovery. What everyone from job seekers to investors to Washington pols wants to know is: Will job growth pick up enough in 2004 to cut unemployment and keep consumers spending? After all, companies have become adept at lifting the output of their existing workers, causing productivity to soar in recent quarters.

Rest easy. More jobs are on the way, and the reason lies in the normal ebb and flow of the business cycle. First of all, demand is accelerating. Over the past four quarters, demand has grown 3.8%, up from 2.5% a year ago and from 0.9% two years ago. Faced with unexpectedly strong demand this summer, businesses had to fatten payrolls by 125,000 in September and 126,000 in October. Better yet, several key indicators point to further job gains.

Second, and equally important, the unusually rapid rate of productivity is set to slow, perhaps sharply. That means economic growth next year will be better balanced between new jobs and productivity gains.


  A hiring rebound could prove pivotal to the 2004 Presidential election, but only if the turnaround carries into the factory sector, the source of all job losses over the past year. That's because most of the big manufacturing states are also major sources of electoral votes.

There's no question that the long-term trend in productivity growth has shifted up in recent years, which is helping to lift profits, real wages, and living standards. However, in the short run, productivity is highly sensitive to the up-and-down movements of the economy.

For example, early in a recovery, productivity typically soars. After working to restore profit margins by shedding workers, businesses are hesitant to take on the added cost of new workers when demand picks up, fearing the uptick won't last. Plus, the workers that companies hang on to tend to be the most productive. The result: Productivity jumps, as in the second and third quarters, when it shot up by 7% and 8.1%, respectively.


  But even with the fanciest of technologies, businesses cannot extract those kinds of productivity gains over the long haul, since there are limits. Like a runner in a marathon, productivity can pick up the pace for a short time, but it will have to give back some of the gains later. That's exactly what happened in the jobless recovery after the 1990-91 recession: Yearly growth in productivity plummeted from 4% in 1992 to zero in 1993, when demand zoomed ahead and businesses began to hire.

A similar short-term pattern is emerging now. Keep in mind that productivity is output per hour worked. According to the employment report, over the past 11 quarters, the work-hours denominator has fallen because of shrinking payrolls and a shorter workweek for those remaining. So even the small gains in output caused a big boost in productivity.

But for the first time since before the recession began, hours worked are rising. In October, they were up at an annual rate of 2.2% from the third quarter. Further gains in November and December will push the pace higher. At the same time, real gross domestic product is slowing after last quarter's 7.2% surge, which reflected temporary boosts. As a result, productivity this quarter will struggle to grow.


  Productivity growth won't have to slip all the way to zero, as it did in 1993, to ensure solid job growth. If the economy grows by 4% in the coming year, derived from, say, 2% productivity growth and 2% job growth, that would translate into job gains averaging about 200,000 per month. That's not an irrational expectation. Payroll gains in the third year of the recovery after the 1990-91 recession averaged 268,000 per month.

Moreover, a pace of 200,000 a month will gradually push down the unemployment rate. Data from the Labor Dept.'s survey of households, from which the unemployment rate is derived, suggest that employment must grow by about 140,000 a month to absorb all the new people looking for work and hold the jobless rate steady. A year of job gains averaging 200,000 per month would cut the rate by about half a percentage point.

Already, labor market indicators that foreshadow job growth strongly suggest bigger payroll gains are on the way. Through October, temporary jobs are up six months in a row, posting the largest gain in 3 1/2 years. The average workweek has registered the largest three-month increase since 1997. And the decline in new claims for unemployment benefits picked up speed in November.


  Even the outlook for manufacturing jobs looks a little brighter. To be sure, U.S. factories face a long-term decline amid global competition, and job outsourcing is becoming more common. But in the short run, stronger demand is generating some bounce in manufacturing, which should generate new jobs in the first half of 2004.

The clearest evidence of a factory rebound comes from the nation's purchasing managers. The Institute for Supply Management's October survey showed the factory sector grew for the fourth month in a row. In addition, the ISM's index of total orders rose to 64.3%, and export demand alone jumped to 59.6%. Both are the highest readings since 1994. Because inventories outside of motor vehicles are so low, producers are not able to fill these orders with warehoused goods. That means nonauto industrial production should continue to grow in the fourth quarter and into 2004.

Hiring in manufacturing lagged behind total payrolls because demand for goods had fallen off dramatically. At the same time, the factory sector was on the leading edge of the productivity boom, powered by industries producing high-tech equipment. Over the past five years, output per hour in the nonfarm business sector rose 3.6% annually, while factory productivity increased by 4.2%. With demand so weak, it's not surprising that 1.3 million factory jobs have disappeared in this recovery. One bit of good news: Last month's loss of 24,000 jobs was the smallest in three years, and layoffs have been declining since the summer.


  Factory workers account for only 11.2% of all employees, so it's easy to dismiss their significance to the economy. But the fate of the factory worker may play an outsize role in the Presidential election. Therefore, the trend in factory jobs will bear scrutiny.

Of the 10 states with the worst unemployment rates in September, half have a higher share of factory-job payrolls than the U.S. average. Those five states control a quarter of the electoral votes needed to win the Presidency. Add in California, where the factory-job ratio is on par with the national rate, and the votes jump to 123, or almost half of the 270 needed.

Politics aside, the job outlook is better now than at any time in the past three years. That's due in large part to demand speeding up and productivity pulling back to a more sustainable level. That combination will assure both rising business profits and increased household income, the keys to keeping the recovery going.

Note: This story originally appeared in the November 24, 2003 issue of BusinessWeek

By James C. Cooper & Kathleen Madigan